RUTGERS UNIVERSITY LIBRARY RESEARCH GUIDE ON ARTIFICIAL INTELLIGENCE:
GLOSSARY OF AI TERMS
Artificial Intelligence
AI refers to computer systems capable of performing tasks that typically require human intelligence, such as reasoning, learning, perception, and language understanding. These systems analyze vast datasets, recognize patterns, and make decisions with unprecedented speed and accuracy.
Machine Learning
Machine learning utilizes data and algorithms to train computers to make classifications, generate predictions, or identify similarities or trends across large datasets.
Natural Language Processing (NLP)
A subset of machine learning that trains computers to understand, interpret, and manipulate human language.
Deep Learning
A subset of machine learning that involves neural networks with many layers. It distinguishes itself from other forms of neural networks primarily through its capability to learn features automatically from data.
Neural Networks
A method to train computers to process data in a way that’s inspired by the human brain, using a layered, interconnected neuron-inspired structure.
Generative Artificial Intelligence
Generative AI models use neural networks to identify the patterns and structures within existing data to generate new and original content.
Large Language Model (LLM)
Large language models (LLMs) are a category of deep learning models trained on immense amounts of data, making them capable of generating natural language and other types of content based on probabilistic methods to perform a wide range of tasks.
Agentic AI
Agentic AI is an artificial intelligence system that can accomplish a specific goal with limited supervision. It consists of AI agents—machine learning models that mimic human decision-making to solve problems in real time.
Artificial General Intelligence
Artificial general intelligence (AGI)—sometimes called human‑level intelligence AI is a type of artificial intelligence that would match or surpass human capabilities across virtually all cognitive tasks.
TAKE THIS LinkedIn COURSE TO UNDERSTAND GENERATIVE AI VS. TRADITIONAL AI (1 hour 17 min. duration)
AI is transforming chemistry through Machine Learning and Generative AI.
| Generative AI | Machine Learning | |
| Primary Goal | Creation and Innovation | Prediction and Decision-Making |
| Core Function | Creates new data/content (e.g. a novel molecule) similar to its training data but not identical | Learns patterns from existing data to make predictions, classifications, or optimizations |
| Output | Novel, original outputs (e.g., a new molecular structure, a synthesis pathway) | Predictive results, classifications, or optimized parameters (e.g., optimized reaction temperature) |
Machine Learning Applications:
Predictive modeling of molecular properties: machine learning, deep learning algorithms, are used to predict a compound's properties, such as its toxicity, solubility, or activity against a biological target, based on its chemical structure. This allows researchers to quickly filter out unpromising candidates from large virtual libraries before expensive lab synthesis.
Process Optimization: Machine learning, including reinforcement learning, can analyze real-time and historical plant data to predict and dynamically adjust process parameters (temperature, pressure, flow rates) to maximize yield, minimize waste, and optimize energy consumption in chemical manufacturing
Materials Science: Machine learning is employed to predict the properties and performance of new materials (plymers battery components) based on their composition and structure, significantly accelerating the discovery of materials with desired characteristics.
Generative AI Applications
Generative AI models are used to design entirely new molecular structures from scratch that are predicted to have specific, desired properties (e.g., high binding affinity to a target protein and low toxicity). This is a shift from screening existing libraries to actively creating optimal compounds.
Reaction Prediction and Retrosynthesis: Generative AI models can predict the product of a chemical reaction given the reactants, or, in the case of retrosynthesis, they can work backward from a target molecule to identify the sequence of chemical reactions and starting materials needed to synthesize it. This helps chemists plan complex synthesis pathways more efficiently.
Virtual Screening and Lead Optimization: After generating novel molecules, generative AI can be used alongside ML to further refine the structures, ensuring they are not only potent but also synthetically feasible and have optimal pharmacokinetic properties.
Watch: The AI Future is Here for Drug Discovery. Source: Bloomberg Podcasts, YouTube, Jan. 22, 2025
How AI is Taking Over Every Step of Drug Discovery
Accurate Structure Prediction of Biomolecular Interactions with AlphaFold3
Mayo Clinic Deploys NVIDIA Blackwell Infrastructure to Drive Generative AI Solutions in Medicine
AI for Science: Six Trends Powering Cutting-Edge Research
Dow Turns to AI to Accelerate Chemical Search
Artificial Intelligence at Dow Chemical - Two Use Cases at the World's Largest Chemical Company
Most academic publishers now require authors to disclose any use of AI tools in article preparation, including the specific tool, version, prompts, and which sections involved AI assistance.
ALWAYS CHECK WITH EACH PUBLISHER'S AI POLICY PRIOR TO SUBMISSION!
Artificial Intelligence (AI) Best Practices and Policies at ACS Publications